List your company
ResearchFormerly WizWand

Thematic fit estimation on Fer-Ins

0.7Spearman CorrelationStep-by-Step Prompting
0.27360.38430.4950.6057Oct 19, 2024
Updated 8mo ago

Evaluation results

MethodLinks
Step-by-Step PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.3.1, Input Form=Lemma Tuples, Output Form=Numeric2024.100.7
Step-by-Step PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.3.2, Input Form=Lemma Tuples, Output Form=Categorical2024.100.68
Step-by-Step PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.4.1, Input Form=Generated Sentences, Output Form=Numeric2024.100.66
Step-by-Step PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.4.2, Input Form=Generated Sentences, Output Form=Categorical2024.100.66
Simple PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.1.2, Input Form=Lemma Tuples, Output Form=Categorical2024.100.65
Simple PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.1.1, Input Form=Lemma Tuples, Output Form=Numeric2024.100.64
GPT4.1 (Exp.3.1)Reasoning=Step-by-Step Prompting, Input=Lemma Tuples, Output=Numeric2024.100.59
GPT4.1 (Exp.3.2)Reasoning=Step-by-Step Prompting, Input=Lemma Tuples, Output=Categorical2024.100.59
GPT4.1 (Exp.1.1)Reasoning=Standard Prompting, Input=Lemma Tuples, Output=Numeric2024.100.56
GPT4.1 (Exp.1.2)Reasoning=Standard Prompting, Input=Lemma Tuples, Output=Categorical2024.100.56
Simple PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.2.2, Input Form=Generated Sentences, Output Form=Categorical2024.100.54
GPT4.1 (Exp.4.1)Reasoning=Step-by-Step Prompting, Input=Generated Sentences, Output=Numeric2024.100.53
GPT4.1 (Exp.4.2)Reasoning=Step-by-Step Prompting, Input=Generated Sentences, Output=Categorical2024.100.53
Simple PromptingBackbone=GPT4-Turbo, Experiment ID=Exp.2.1, Input Form=Generated Sentences, Output Form=Numeric2024.100.49
ResRoFA-MT2024.100.48
ResRoFA-MTMethod ID=B12024.100.48
ResRoFA-MTLabel=B12024.100.48
NN RF2024.100.45
NN RFMethod ID=B02024.100.45
NN RFLabel=B02024.100.45
Exp.2.1Backbone=Qwen2024.100.44
Glove-sharedtuning=tuned, data=20% subset2024.100.43
Glove-shared tuned with 20%MMethod ID=B3c2024.100.43
Exp.1.1Backbone=Qwen2024.100.43
Glove-sharedLabel=B3c, Tuning Status=tuned, Training Subset=20%M2024.100.43
GSD20152024.100.42
GSD2015Method ID=BG2024.100.42
GSD2015Label=BG2024.100.42
Exp.2.2Backbone=Qwen2024.100.38
GPT4.1 (Exp.2.2)Reasoning=Standard Prompting, Input=Generated Sentences, Output=Categorical2024.100.37
Exp.1.2Backbone=Qwen2024.100.37
Exp.4.1Backbone=Qwen2024.100.37
Exp.3.1Backbone=Qwen2024.100.36
Exp.4.2Backbone=Qwen2024.100.36
Glove-sharedtuning=not-tuned2024.100.34
Glove-shared not-tunedMethod ID=B3a2024.100.34
Glove-sharedLabel=B3a, Tuning Status=not-tuned2024.100.34
GPT4.1 (Exp.2.1)Reasoning=Standard Prompting, Input=Generated Sentences, Output=Numeric2024.100.33
Exp.3.2Backbone=Qwen2024.100.32
Glove-sharedtuning=tuned2024.100.29
Glove-shared tunedMethod ID=B3b2024.100.29
Glove-sharedLabel=B3b, Tuning Status=tuned2024.100.29